A theory for gentle but persistent ascent is developed in which a convergent flow is induced by the release of the frictional constraint in the convective boundary layer (CBL) at sunset if a mesoscale warm tongue is present. The ascent mechanism is essentially the Blackadar mechanism that produces nocturnal low-level jets (LLJs) as inertial oscillations, but operating on horizontally inhomogeneous CBLs. We study the problem analytically, with the nightime response obtained as a solution of the linearized Boussinesq equations of motion, thermal energy, and mass conservation on an f
-plane. The motion can be described as an inertia-gravity response to the release of the frictional constraint. The vertical velocity revealed by the solution increases with the amplitude of the buoyancy variation, depth of the CBL, and wavenumber of the buoyancy variation (larger velocity for smaller-scale variations). Stable stratification has a lid effect, with a larger Brunt-Vaisala frequency associated with a smaller vertical velocity. For typical parameter values, the vertical velocity peaks in a 5-10 cm/s range, with parcels rising 0.5-1 km over the ascent phase of the oscillation, which lasts 6-8 h.
PECAN data were used as a qualitative check on our hypothesis that the same mechanism that produces LLJs from CBLs can lead to gentle but persistent ascent if a mesoscale warm tongue is present. We discuss four cases in which Blackadar-like LLJs, persistent weak rising motion, and pristine CI developed at night after an afternoon in which conditions roughly conformed to the restrictions of our theory: 1 June (first day of PECAN), 2 June (IOP1), 14 June (UFO2), and 5 July (IOP19). Since vertical velocities on the order of the anticipated ascent rates (generally less than 10 cm/s) are less than the velocity precisions of the lidars and profilers deployed in PECAN, they cannot be verified directly using those instruments. Instead, we verified gentle persistent vertical velocities indirectly, by tracking the motion of coherent layered structures in time-height plots of data from the UMBC CL31 ceilometer at the S-Pol site, and data from the NCAR/EOL 449 MHz profiler, Millersville University MPL, and NCAR/EOL WV-DIAL at the Ellis site, whenever such structures were evident.